Abstract
Background
Antimicrobial resistance is driven by inappropriate antibiotic use. In Cambodia, young children frequently receive antibiotics despite declining rates of common illnesses. We analyzed nationally representative data to describe trends and determinants of antibiotic use for acute respiratory infection (ARI), fever, and diarrhea among under-five children.
Materials and Methods
We performed a secondary analysis of the Cambodia Demographic and Health Surveys from 2010, 2014, and 2021–2022. Caregivers reported whether children (<60 months) experienced ARI, fever, or diarrhea in the two weeks preceding the survey and whether antibiotics were administered. We calculated weighted prevalence estimates of each illness and the proportion of episodes treated with antibiotics. For fever and diarrhea separately, we fitted multivariable logistic regression models—stratified by survey year—to identify associations between antibiotic receipt and child age, sex, urban vs. rural residence, household wealth quintile, and maternal education, accounting for sampling weights and clustering.
Results
Between 2010 and 2021–2022, the two-week prevalence of fever declined from 26.8% to 13.2% (P<0.001) and diarrhea from 14.5% to 6.9% (P<0.001), yet antibiotic use remained high. In 2021–2022, 53% of fever episodes and 29% of diarrheal episodes were treated with antibiotics. Urban residence was associated with higher odds of antibiotic use for fever (adjusted odds ratio [aOR], 1.85; 95% confidence interval [CI], 1.05–3.25) and diarrhea (aOR, 2.38; 95% CI, 1.02–5.54) vs. rural residence. In 2010, children in the richest quintile had greater odds of antibiotic use for fever (aOR, 2.10; 95% CI, 0.89–4.97) and diarrhea (aOR, 3.39; 95% CI, 1.61–7.11) compared to those in the poorest quintile; these disparities persisted, though attenuated, in later surveys.
Conclusion
Despite substantial declines in childhood illness, antibiotic use among Cambodian children under five remains disproportionately high—especially in urban and wealthier households. Antimicrobial stewardship initiatives should focus on reducing overuse in privileged populations while ensuring equitable access in underserved areas.
Keywords: Pediatric, Antibiotic, Consumption, DHS
Graphical Abstract
Introduction
Antimicrobial resistance (AMR), a growing global health threat, is significantly driven by antibiotic overuse. This issue is particularly critical in low- and middle-income countries like Cambodia, where young children frequently receive antibiotics for common illnesses, often without medical necessity [1,2]. Data from 2014, showing nearly 83% of Cambodian children under five with respiratory infection symptoms received antibiotics, illustrates potential overuse in this vulnerable group [3]. Such patterns highlight the public health imperative to improve antibiotic use and strengthen antimicrobial stewardship to combat emerging resistance.
Specific factors within Cambodia contribute to these high rates of antibiotic use. Despite significant economic growth and improvements in child health over the past decade, regulatory oversight of pharmaceuticals remains limited [4]. Compounding this, antibiotics are readily available without prescription, as over half of healthcare services are delivered by private providers, alongside uncontrolled pharmacy sales [5]. This combination of easy access and weak regulation fuels pervasive community antibiotic consumption, intensifying concerns about AMR and its consequences.
To address this knowledge gap, we analyzed Cambodia Demographic and Health Survey (CDHS) data (2010–2022) to examine antibiotic use trends and associated factors in children under five. Using three survey rounds, this analysis provides a comprehensive overview to inform public health interventions and guide antimicrobial stewardship for promoting judicious antibiotic use in Cambodia.
Materials and Methods
1. Study design and data source
We analyzed cross-sectional data from three rounds of the CDHS conducted in 2010, 2014, and 2021–2022 [6,7,8]. Each survey used a stratified two-stage cluster sampling design to obtain a nationally representative sample of households with under-five children. In each selected cluster, households were randomly chosen, and all living children under 5 years in those households were included in the analysis, provided that data on recent illnesses and antibiotic treatments were available. Data were collected through structured questionnaires administered to mothers or primary caregivers of the children. All three CDHS rounds used face-to-face, household interviews. Fieldwork months were: July 23, 2010–January 20, 2011 (2010 CDHS), June 2–November 14, 2014 (2014 CDHS), and September 15, 2021–February 15, 2022 (2021–2022 CDHS). Because the rounds covered different calendar months (including both rainy and dry seasons), residual seasonality cannot be ruled out.
2. Ethics statement
The CDHS protocols were approved by institutional review boards in Cambodia and the DHS Program, and informed consent was obtained from participants (IRB No. 083 NECHR). This study constitutes a secondary analysis of de-identified, publicly available DHS data and was therefore exempt from further ethical review.
3. Case definitions and variables
We focused on three common pediatric illness syndromes based on caregiver recall in the 2 weeks prior to the survey: acute respiratory infection (ARI) symptoms, fever, and diarrhea. ARI was defined according to WHO Integrated Management of Childhood Illness criteria for presumptive pneumonia [9]. Fever (febrile illness) was defined as caregiver-reported fever in the preceding two weeks; thermometer measurements and temperature thresholds (e.g., ≥38°C) were not collected by DHS. Diarrhea was defined as an episode of abnormally frequent loose or liquid stools, and cases with blood in stool were classified as dysentery. For each illness episode reported, we recorded whether the child received an antibiotic treatment.
4. Statistical analysis
We first performed descriptive analyses for each survey year to examine changes over time in sample characteristics, illness prevalence (ARI, fever, diarrhea), care-seeking rates, and the proportion of illnesses treated with antibiotics. Trends over 2010–2022 in disease prevalence and antibiotic use were visualized in figures, and differences in categorical variables across survey rounds were tested using chi-square tests. We then conducted multivariable logistic regression separately for each survey year to identify factors associated with a child receiving antibiotics for an illness. The analyses were restricted to children who had experienced ARI, fever, or diarrhea in the preceding two weeks. Predictor variables in the models included the child’s sex, age group (in months), urban vs. rural residence, household wealth index quintile, and mother’s education level. Odds ratios (OR) with 95% confidence intervals (CIs) were estimated for each factor. Because very few ARI cases were reported, we excluded ARI from the regression analysis to ensure adequate sample size for reliable estimates. We stratified regressions by survey year to allow associations to vary over time rather than assume constant effects across 2010–2022; although this reduces the sample size per model, the large survey samples still provide adequate precision for key estimates. All analyses accounted for the complex survey design by applying sample weights and adjusting for clustering, yielding nationally representative estimates. Statistical significance was defined as a two-sided P-value <0.05 (or 95% CI not overlapping 1.0). Data were managed and analyzed using SAS software (version 9.4, SAS Institute, Cary, NC, USA).
Results
1. Population characteristics
Each survey round included approximately 7,000–8,000 children under five years of age (Table 1). The basic demographic profile of the under-five population was similar across 2010, 2014, and 2021–2022. There were no significant differences in the sex ratio or age distribution of the children between survey years (Table 1). Notably, the residence and socioeconomic makeup shifted over time. In 2010 and 2014, about 72–74% of under-five children lived in rural areas, compared to about 66% in 2021–2022, reflecting increasing urbanization. The wealth distribution also changed: in 2010, 25% of the children were from households in the poorest wealth quintile and 21% were from the richest quintile, whereas by 2021–2022 the poorest quintile comprised 30% of the under-five population and the richest quintile 14%. Maternal education levels improved over the study period; for example, the proportion of mothers with no formal education declined from 21% in 2010 to 13% in 2021–2022, alongside a rise in mothers with at least a secondary education. These shifts in residence, wealth, and maternal education were statistically significant across survey rounds (P<0.01).
Table 1. Background characteristics of children under five and their households in Cambodia, DHS 2010, 2014, and 2021–2022.
| Variables | 2010 No. (%) | 2014 No. (%) | 2020–2021 No. (%) | P-value | |
|---|---|---|---|---|---|
| U5 population | 7,820 (100) | 6,971 (100) | 8,025 (100) | ||
| Sex | 0.8955 | ||||
| Male | 3,987 (51.0) | 3,496 (50.2) | 4,085 (50.9) | ||
| Female | 3,833 (49.0) | 3,475 (49.9) | 3,940 (49.1) | ||
| Age (months) | 0.0728 | ||||
| <6 | 723 (9.3) | 703 (10.1) | 805 (10.1) | ||
| 6–11 | 821 (10.5) | 762 (10.9) | 846 (10.6) | ||
| 12–23 | 1,619 (20.7) | 1,441 (20.7) | 1,695 (21.3) | ||
| 24–35 | 1,594 (20.4) | 1,386 (19.9) | 1,494 (18.7) | ||
| 36–47 | 1,578 (20.2) | 1,306 (18.7) | 1,618 (20.3) | ||
| 48–59 | 1,485 (19.0) | 1,373 (19.7) | 1,503 (18.9) | ||
| Residence | <0.0001 | ||||
| Urban | 2,053 (26.3) | 1,923 (27.6) | 2,699 (33.6) | ||
| Rural | 5,767 (73.8) | 5,048 (72.4) | 5,326 (66.4) | ||
| Wealth index | <0.0001 | ||||
| Poorest | 1,977 (25.3) | 1,634 (23.4) | 2,420 (30.2) | ||
| Poorer | 1,540 (19.7) | 1,298 (18.6) | 1,460 (18.2) | ||
| Middle | 1,298 (16.6) | 1,105 (15.9) | 1,435 (17.9) | ||
| Richer | 1,355 (17.3) | 1,211 (17.4) | 1,552 (19.3) | ||
| Richest | 1,650 (21.1) | 1,723 (24.7) | 1,158 (14.4) | ||
| Maternal education | <0.0001 | ||||
| No education | 1,670 (21.4) | 985 (14.1) | 1,071 (13.4) | ||
| Primary | 4,110 (52.6) | 3,447 (49.5) | 3,394 (42.3) | ||
| Secondary | 1,903 (24.3) | 2,274 (32.6) | 3,126 (39.0) | ||
| Higher | 137 (1.8) | 265 (3.8) | 434 (5.4) | ||
| Household members (no.) | <0.0001 | ||||
| 1–3 | 799 (10.2) | 794 (11.4) | 1,037 (12.9) | ||
| 4–6 | 4,495 (57.5) | 4,078 (58.5) | 5,385 (67.1) | ||
| 7–10 | 2,172 (27.8) | 1,798 (25.8) | 1,454 (18.1) | ||
| >11 | 354 (4.5) | 301 (4.3) | 149 (1.9) | ||
Prevalence of acute respiratory infection (ARI), fever, and diarrhea among children under five by urban–rural residence in Cambodia, 2010, 2014, and 2021–2022 demographic and health survey. In each survey year, rural children had a higher prevalence of all three illnesses compared to urban children. The prevalence of ARI, fever, and diarrhea declined from 2010 to 2021–2022 in both urban and rural populations.
Data source: Cambodia Demographic and Health Surveys (2010, 2014, 2021–2022).
DHS, Demographic and Health Survey.
2. Prevalence of ARI, fever, and diarrhea
The prevalence of the three monitored childhood illnesses remained relatively stable between 2010 and 2014, then declined markedly by 2021–2022 (Fig. 1). In 2010, 6.1% of children under five had symptoms of ARI in the previous two weeks, compared to 5.2% in 2014 (a difference that was not statistically significant, P>0.05). By 2021–2022, reported ARI prevalence had decreased to 1.7% of children (a significant drop, P<0.001). A similar trend was observed for fever: approximately 26.8% of children had a fever in the two weeks before the survey in 2010, essentially the same as 26.0% in 2014, but this proportion fell to 13.2% in 2021–2022. The prevalence of diarrhea was 14.5% in 2010 and 12.3% in 2014, and then dropped by about half to 6.9% in 2021–2022. All the declines in illness prevalence observed between 2014 and 2021–2022 were statistically significant (Chi-square test P<0.001). Supplementary Table 1 provides the detailed prevalence values for each survey year. Rural children had a higher prevalence of all three illnesses than urban children, but both urban and rural populations experienced these significant downward trends in disease burden (Fig. 1).
Figure 1. Prevalence of acute respiratory infection (ARI), fever, and diarrhea among children under five by urban–rural residence in Cambodia, 2010, 2014, and 2021–2022 Demographic and Health Surveys.
In each survey year, rural children had a higher prevalence of all three illnesses compared to urban children. The prevalence of (A) ARI, (B) fever, and (C) diarrhea declined from 2010 to 2021–2022 in both urban and rural populations.
Data source: Cambodia Demographic and Health Surveys (2010, 2014, 2021–2022).
3. Antibiotic use patterns in childhood illness
Antibiotic treatment for common pediatric illnesses remained widespread across all survey years (for proportions by year and residence). Despite the marked decline in illness prevalence between 2010 and 2021–2022, the share of episodes treated with antibiotics remained high, particularly for fever (~53% in 2021–2022). Fever was the most frequent indication for antibiotic use, followed by diarrhea. Tables 2 and 3 present multivariable associations between child and household characteristics and antibiotic use for fever and diarrhea, respectively. For fever (Table 2), no consistent associations were found with child sex or age group across survey rounds. However, children from urban areas were significantly more likely to receive antibiotics in 2021–2022 (adjusted OR, 1.85; 95% CI, 1.05–3.25, vs. rural), reversing a pattern seen in 2010. Wealth-related disparities were also evident: in 2010, children from the richest quintile had over twice the odds of receiving antibiotics for fever compared to the poorest group (OR, 2.10; 95% CI, 0.89–4.97), though these differences attenuated in later years. For diarrhea (Table 3), similar patterns were observed. Children from rural households were less likely to receive antibiotics than those from urban households in 2021–2022 (OR, 0.42; 95% CI, 0.18–0.98). Socioeconomic status again emerged as a key determinant: in 2010, children in the richest quintile had over threefold higher odds of receiving antibiotics for diarrhea compared to the poorest quintile (OR, 3.39; 95% CI, 1.61–7.11).
Table 2. Association of antibiotic use for fever among children under five in Cambodia (2010 / 2014 / 2021–2022).
| Variables | 2010 | 2014 | 2020–2021 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No. (%) | OR | 95% CI | No. (%) | OR | 95% CI | No. (%) | OR | 95% CI | ||
| Sex | ||||||||||
| Male | 26 (55.3) | Ref | 6 (40.0) | Ref | 4 (40.0) | Ref | ||||
| Female | 21 (44.7) | 0.84 | (0.47, 1.50) | 9 (60.0) | 1.69 | (0.60, 4.76) | 6 (60.0) | 1.72 | (0.48, 6.11) | |
| Age (months) | ||||||||||
| <6 | 6 (12.8) | Ref | 1 (6.7) | Ref | 0 | Ref | ||||
| 6–11 | 9 (19.2) | 0.64 | (0.22, 1.84) | 3 (20) | 1.55 | (0.16, 15.07) | 1 (10.0) | |||
| 12–23 | 14 (29.8) | 0.56 | (0.21, 1.49) | 6 (40) | 1.57 | (0.19, 13.14) | 1 (10.0) | |||
| 24–35 | 9 (19.2) | 0.46 | (0.16, 1.31) | 1 (6.7) | 0.31 | (0.02, 4.94) | 1 (10.0) | |||
| 36–47 | 7 (14.9) | 0.41 | (0.14, 1.25) | 3 (20.0) | 1.27 | (0.13, 12.30) | 5 (50.0) | |||
| 48–59 | 2 (4.3) | 0.14 | (0.03, 0.70) | 1 (6.7) | 0.43 | (0.03, 6.88) | 2 (20.0) | |||
| Residence | ||||||||||
| Urban | 10 (21.3) | Ref | 1 (6.7) | Ref | 6 (60.0) | Ref | ||||
| Rural | 37 (78.7) | 1.16 | (0.57, 2.34) | 14 (93.3) | 4.9 | (0.64, 37.34) | 4 (40.0) | 0.27 | (0.07, 0.95) | |
| Wealth index | ||||||||||
| Poorest | 9 (19.2) | Ref | 4 (26.7) | Ref | 3 (30.0) | Ref | ||||
| Poorer | 7 (14.9) | 0.96 | (0.35, 2.59) | 6 (40.0) | 2.16 | (0.61, 7.72) | 2 (20.0) | 0.94 | (0.16, 5.68) | |
| Middle | 7 (14.9) | 1.14 | (0.42, 3.08) | 1 (6.7) | 0.38 | (0.04, 3.41) | 0 | |||
| Richer | 11 (23.4) | 1.69 | (0.69, 4.11) | 3 (20) | 1.12 | (0.25, 5.02) | 4 (40.0) | 2.41 | (0.53, 10.89) | |
| Richest | 13 (27.7) | 2.1 | (0.89, 4.97) | 1 (6.7) | 0.27 | (0.03, 2.40) | 1 (10.0) | 1.37 | (0.14, 13.37) | |
| Maternal education | ||||||||||
| No education | 9 (19.2) | Ref | 1 (6.7) | Ref | 1 (10.0) | Ref | ||||
| Primary | 26 (55.3) | 1.12 | (0.52, 2.40) | 6 (40.0) | 1.54 | (0.19, 12.86) | 6 (60.0) | 2.27 | (0.27, 19.01) | |
| Secondary | 10 (21.3) | 1.08 | (0.44, 2.69) | 7 (46.7) | 3.05 | (0.37, 24.88) | 3 (30.0) | 1.39 | (0.14, 13.45) | |
| Higher | 2 (4.3) | 3.61 | (0.74, 17.52) | 1 (6.7) | 5.68 | (0.35, 92.46) | 0 | |||
| Household members | ||||||||||
| 1–3 | 9 (19.2) | Ref | 2 (13.3) | Ref | 0 | Ref | ||||
| 4–6 | 21 (44.7) | 0.46 | (0.21, 1.02) | 11 (73.3) | 8 (80.0) | |||||
| 7–10 | 16 (34) | 0.78 | (0.34, 1.80) | 2 (13.3) | 2 (20.0) | |||||
| >11 | 1 (2.1) | 0.38 | (0.05, 3.09) | 0 | 0 | |||||
2010, 2014, and 2021–2022 refer to Cambodia Demographic and Health Survey (DHS) years. Values are weighted percentages of children under five (unless stated otherwise). Wealth index is a relative household wealth measure provided by DHS, divided into quintiles from poorest to richest.
OR, odds ratio; CI, confidence interval.
Table 3. Association of antibiotic use for diarrhea among children under five in Cambodia (2010 / 2014 / 2021–2022).
| Variables | 2010 | 2014 | 2020–2021 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No. (%) | OR | 95% CI | No. (%) | OR | 95% CI | No. (%) | OR | 95% CI | ||
| Sex | ||||||||||
| Male | 39 (50.7) | Ref | 17 (50.0) | Ref | 10 (43.5) | Ref | ||||
| Female | 38 (49.3) | 1.15 | (0.73, 1.83) | 17 (50.0) | 1.18 | (0.59, 2.34) | 13 (56.5) | 1.61 | (0.69, 3.73) | |
| Age (months) | ||||||||||
| <6 | 9 (11.7) | Ref | 4 (11.8) | Ref | 0 | Ref | ||||
| 6–11 | 15 (19.5) | 0.72 | (0.30, 1.70) | 5 (14.7) | 0.64 | (0.17, 2.45) | 3 (13.6) | |||
| 12–23 | 26 (33.8) | 0.83 | (0.38, 1.84) | 11 (32.4) | 0.77 | (0.24, 2.49) | 7 (31.8) | |||
| 24–35 | 13 (16.9) | 0.63 | (0.26, 1.53) | 4 (11.8) | 0.41 | (0.10, 1.69) | 7 (31.8) | |||
| 36–47 | 9 (11.7) | 0.63 | (0.24, 1.66) | 6 (17.7) | 0.99 | (0.27, 3.64) | 4 (18.2) | |||
| 48–59 | 5 (6.5) | 0.41 | (0.13, 1.27) | 4 (11.8) | 0.8 | (0.19, 3.29) | 1 (4.6) | |||
| Residence | ||||||||||
| Urban | 20 (26.0) | Ref | 6 (17.7) | Ref | 10 (43.5) | Ref | ||||
| Rural | 57 (74.0) | 0.71 | (0.42, 1.21) | 28 (82.3) | 1.52 | (0.62, 3.73) | 13 (56.5) | 0.42 | (0.18, 0.98) | |
| Wealth index | ||||||||||
| Poorest | 12 (15.6) | Ref | 8 (23.5) | Ref | 11 (47.8) | Ref | ||||
| Poorer | 13 (16.9) | 1.36 | (0.61, 3.04) | 12 (35.3) | 2.41 | (0.96, 6.02) | 3 (13.0) | 0.5 | (0.14, 1.81) | |
| Middle | 12 (15.6) | 1.65 | (0.73, 3.75) | 1 (2.9) | 0.25 | (0.03, 2.01) | 1 (4.4) | 0.27 | (0.03, 2.13) | |
| Richer | 20 (26.0) | 3.01 | (1.44, 6.31) | 7 (20.6) | 1.74 | (0.62, 4.92) | 5 (21.7) | 1.12 | (0.38, 3.31) | |
| Richest | 20 (26.0) | 3.39 | (1.61, 7.11) | 6 (17.7) | 1.08 | (0.37, 3.16) | 3 (13.0) | 1.31 | (0.35, 4.88) | |
| Maternal education | ||||||||||
| No education | 11 (14.3) | Ref | 2 (5.9) | Ref | 2 (8.7) | Ref | ||||
| Primary | 45 (58.4) | 1.94 | (0.99, 3.81) | 19 (55.9) | 2.8 | (0.64, 12.14) | 14 (60.9) | 2.24 | (0.50, 10.08) | |
| Secondary | 18 (23.4) | 2.17 | (1.00, 4.70) | 10 (29.4) | 2.4 | (0.52, 11.13) | 5 (21.7) | 1.27 | (0.24, 6.66) | |
| Higher | 3 (3.9) | 9.28 | (2.16, 39.85) | 3 (8.8) | 8.02 | (1.27, 50.70) | 2 (8.7) | 5.25 | (0.69, 40.04) | |
| Household members | ||||||||||
| 1–3 | 14 (18.2) | Ref | 3 (8.8) | Ref | 2 (8.7) | Ref | ||||
| 4–6 | 39 (50.7) | 0.52 | (0.27, 0.99) | 25 (73.5) | 2.13 | (0.63, 7.17) | 14 (60.9) | 1.46 | (0.32, 6.54) | |
| 7–10 | 20 (26.0) | 0.58 | (0.28, 1.18) | 5 (14.7) | 0.87 | (0.20, 3.70) | 7 (30.4) | 2.75 | (0.55, 13.62) | |
| >11 | 4 (5.2) | 0.75 | (0.23, 2.39) | 1 (2.9) | 0.79 | (0.08, 7.74) | 0 | |||
| Illness episodes in past 2 weeks (unweighted counts) | ||||||||||
| Acute respiratory illness | 475 | 364 | 139 | |||||||
| Fever | 2,096 | 1,813 | 1,058 | |||||||
| Diarrhea | 1,135 | 855 | 551 | |||||||
2010, 2014, and 2021–2022 refer to Cambodia Demographic and Health Survey (DHS) years. Values are weighted percentages of children under five (unless stated otherwise). Wealth index is a relative household wealth measure provided by DHS, divided into quintiles from poorest to richest. Episode counts are unweighted; prevalence values in the text are weighted.
OR, odds ratio; CI, confidence interval.
Discussion
Our most important finding is the marked decline in the prevalence of common childhood illnesses in Cambodia between 2010 and 2022. These declines align with broader trends in low- and middle-income countries and are consistent with improvements in vaccination, sanitation, and child health services. These declines align with broader trends observed in many low- and middle-income countries, likely reflecting improvements in vaccination coverage, sanitation, and other public health interventions [10]. In particular, the national roll-out of measles-rubella containing vaccines, alongside expansion of child health programs, may have corresponded with substantial reductions in ARI/febrile events and diarrheal disease burden [11]. Our results are consistent with studies from other Southeast Asian settings that document significant progress in reducing pediatric diarrheal illness and respiratory infections in the past decade, underscoring Cambodia’s achievements within this regional context [12,13]. Notably, the 2021–2022 CDHS was conducted during the coronavirus disease 2019 (COVID-19) pandemic, when public health measures and reduced social mixing may have suppressed common infections and affected care-seeking behavior, potentially contributing to the lower illness prevalence and antibiotic use patterns observed.
Despite the decline in illnesses, our analysis found that antibiotic use in children remained high and showed striking disparities by socioeconomic status and residence. Children from wealthier families and those living in urban areas were significantly more likely to receive antibiotics for recent illness episodes than their poorer, rural counterparts, a pattern aligning with multi-country findings that antibiotic treatment rates in childhood are lowest among the most disadvantaged groups. In Cambodia’s context, relatively unrestricted access to antibiotics through private pharmacies and informal drug sellers may enable easier antibiotic procurement for urban and affluent households [14,15]. Socioeconomic factors are known to strongly influence antibiotic-seeking and prescribing behaviors in low-resource settings, raising concerns that better-off populations might be overusing antibiotics while children in underserved communities face barriers to accessing appropriate treatment [16]. Importantly, the proportion of illnesses treated with antibiotics changed little over time, indicating that antibiotic use remained high in 2021–2022 despite the substantial drop in illness prevalence. These disparities pose a dual challenge: potential over-prescription in settings with easy access to antibiotics may contribute to AMR, while inadequate access in underserved areas risks under-treatment of bacterial infections; stewardship should promote appropriate antibiotic use while improving equitable access. Thus, stewardship efforts must curb inappropriate antibiotic use in well-served populations even as access to essential treatment is improved for underserved groups. These patterns are broadly consistent with reports from other Southeast Asian settings documenting high pediatric antibiotic exposure and easy retail access to antibiotics in community and private-sector channels. While definitions and data sources differ, Cambodia’s levels appear comparatively high within this regional context.
Several limitations of this study should be acknowledged. First, the illness data were based on caregivers’ recall of symptoms in the prior two weeks, without any clinical confirmation. Because the 2021–2022 round was fielded during the COVID-19 pandemic, pandemic-related behavior changes and health-service disruptions may have affected both reported illness prevalence and antibiotic use, limiting direct comparability with earlier rounds. Reporting of antibiotic use may vary by socioeconomic status or education, with caregivers in poorer or less-educated households potentially under-recognizing or under-recalling antibiotics, which could attenuate or distort observed disparities. Second, we lacked information on specific pathogens, clinical diagnoses, or antibiotic dosing and duration, preventing us from assessing whether reported antibiotic use was appropriate. The DHS did not collect microbiological or clinical data, so we cannot differentiate, for example, bacterial pneumonia cases that required antibiotics from viral respiratory infections that would have self-resolved without antibiotics. Third, the analysis is cross-sectional in nature; associations observed should not be interpreted as causal, and unmeasured temporal factors (such as seasonal differences between survey rounds) could have influenced the observed trends. Additionally, antibiotic use was determined by maternal report, which may be subject to reporting errors or lack of caregiver awareness about the medications given. Moreover, pandemic-era behaviors and service constraints during 2021–2022 may have temporarily altered infection patterns and care-seeking, affecting both reported illness and antibiotic use in that round.
However, the study has several notable strengths. It draws on nationally representative DHS datasets spanning 2010–2022, with a combined sample of nearly 30,000 children under five. The DHS’s standardized sampling and survey procedures ensure comparability across survey rounds, enabling a reliable analysis of trends over the decade. This large, multi-round, population-based dataset provides a comprehensive picture of antibiotic use and illness patterns at the community level, rather than relying solely on facility-based data.
In summary, this study highlights a disconnect between declining childhood illness prevalence and persistently high antibiotic utilization in Cambodia. In effect, while fewer children are falling ill, antibiotic treatments remain widespread and unevenly distributed, underscoring the need for improved antimicrobial stewardship. National efforts should prioritize ensuring appropriate, evidence-based antibiotic use—particularly in urban and private-sector settings—while safeguarding equitable access for children in poorer and rural communities. At the same time, it is crucial to ensure that children in poorer and rural areas have equitable access to life-saving antibiotics when they are truly needed, in line with the country’s goals of promoting rational antimicrobial use. Strengthening public health interventions that prevent infections alongside educating healthcare providers and communities about judicious antibiotic use will be key to addressing this issue.
ACKNOWLEDGMENT
The authors would like to acknowledge the staff of the Department of Health Services Program, and the National Pediatric Hospital (NPH) in Cambodia, as well as the staff of Korea University College of Medicine and Korea University Anam Hospital, for their valuable support and contribution to this research.
Footnotes
Funding: None.
Conflict of Interest: No conflict of interest.
- Conceptualization: YJC, JEP.
- Data Curation: EH, JS.
- Formal Analysis: EH, JS.
- Funding Acquisition: YJC.
- Investigation: YHL.
- Methodology: YJC, JEP, YSK.
- Project Administration: YJC.
- Supervision: YJC, YSK. Writing –.
- Original Draft: EH, JS. Writing –.
- Review & Editing: YHL, YSK, JEP, YJC.
SUPPLEMENTARY MATERIAL
Proportion of under-five illness episodes treated with antibiotics in Cambodia, by DHS year (2010, 2014, 2021–2022) and residence (urban/rural) for fever, ARI, and diarrhea
References
- 1.Miyazaki A, Tung R, Taing B, Matsui M, Iwamoto A, Cox SE. Frequent unregulated use of antibiotics in rural Cambodian infants. Trans R Soc Trop Med Hyg. 2020;114:401–407. doi: 10.1093/trstmh/traa020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Om C, Daily F, Vlieghe E, McLaughlin JC, McLaws ML. Pervasive antibiotic misuse in the Cambodian community: antibiotic-seeking behaviour with unrestricted access. Antimicrob Resist Infect Control. 2017;6:30. doi: 10.1186/s13756-017-0187-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ardillon A, Ramblière L, Kermorvant-Duchemin E, Sok T, Zo AZ, Diouf JB, Long P, Lach S, Sarr FD, Borand L, Cheysson F, Collard JM, Herindrainy P, de Lauzanne A, Vray M, Delarocque-Astagneau E, Guillemot D, Huynh BT, BIRDY study group Inappropriate antibiotic prescribing and its determinants among outpatient children in 3 low- and middle-income countries: a multicentric community-based cohort study. PLoS Med. 2023;20:e1004211. doi: 10.1371/journal.pmed.1004211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yang D, Plianbangchang P, Visavarungroj N, Rujivipat S. Quality of pharmaceutical items available from drugstores in Phnom Penh, Cambodia. Southeast Asian J Trop Med Public Health. 2004;35:741–747. [PubMed] [Google Scholar]
- 5.Om C, Vlieghe E, McLaughlin JC, Daily F, McLaws ML. Antibiotic prescribing practices: a national survey of Cambodian physicians. Am J Infect Control. 2016;44:1144–1148. doi: 10.1016/j.ajic.2016.03.062. [DOI] [PubMed] [Google Scholar]
- 6.National Institute of Statistics, Directorate General for Health and ICF International. 2014 Cambodia Demographic and Health Survey Key Findings. Rockville, Maryland, USA: National Institute of Statistics, Directorate General for Health and ICF International; 2015. [Google Scholar]
- 7.National Institute of Statistics. Directorate General for Health, and ICF Macro, Cambodia Demographic and Health Survey 2010. Phnom Penh, Cambodia and Calverton, Maryland, USA: National Institute of Statistics, Directorate General for Health, and ICF Macro; 2011. [Google Scholar]
- 8.National Institute of Statistics (NIS) [Cambodia], Ministry of Health (MoH) [Cambodia], and ICF. Cambodia Demographic and Health Survey 2021–22 Final Report. Phnom Penh, Cambodia, and Rockville, Maryland, USA: NIS, MoH, and ICF; 2023. [Google Scholar]
- 9.Gera T, Shah D, Garner P, Richardson M, Sachdev HS. Integrated management of childhood illness (IMCI) strategy for children under five. Cochrane Database Syst Rev. 2016;2016:CD010123. doi: 10.1002/14651858.CD010123.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Black RE, Perin J, Yeung D, Rajeev T, Miller J, Elwood SE, Platts-Mills JA. Estimated global and regional causes of deaths from diarrhoea in children younger than 5 years during 2000-21: a systematic review and Bayesian multinomial analysis. Lancet Glob Health. 2024;12:e919–28. doi: 10.1016/S2214-109X(24)00078-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Knapp JK, Mariano KM, Pastore R, Grabovac V, Takashima Y, Alexander JP, Jr, Reef SE, Hagan JE. Progress toward rubella elimination - Western Pacific Region, 2000-2019. MMWR Morb Mortal Wkly Rep. 2020;69:744–750. doi: 10.15585/mmwr.mm6924a4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.2021 Diarrhoeal Diseases Collaborators. Global, regional, and national age-sex-specific burden of diarrhoeal diseases, their risk factors, and aetiologies, 1990-2021, for 204 countries and territories: a systematic analysis for the Global Burden of Disease Study 2021. Lancet Infect Dis. 2025;25:519–536. doi: 10.1016/S1473-3099(24)00691-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.McAllister DA, Liu L, Shi T, Chu Y, Reed C, Burrows J, Adeloye D, Rudan I, Black RE, Campbell H, Nair H. Global, regional, and national estimates of pneumonia morbidity and mortality in children younger than 5 years between 2000 and 2015: a systematic analysis. Lancet Glob Health. 2019;7:e47–57. doi: 10.1016/S2214-109X(18)30408-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Khan M, Rahman-Shepherd A, Bory S, Chhorn S, Durrance-Bagale A, Hasan R, Heng S, Phou S, Prien C, Probandari A, Saphonn V, Suy S, Wiseman V, Wulandari LPL, Hanefeld J. How conflicts of interest hinder effective regulation of healthcare: an analysis of antimicrobial use regulation in Cambodia, Indonesia and Pakistan. BMJ Glob Health. 2022;7:e008596. doi: 10.1136/bmjgh-2022-008596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lim JM, Chhoun P, Tuot S, Om C, Krang S, Ly S, Hsu LY, Yi S, Tam CC. Public knowledge, attitudes and practices surrounding antibiotic use and resistance in Cambodia. JAC Antimicrob Resist. 2021;3:dlaa115. doi: 10.1093/jacamr/dlaa115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Fernandes Antunes A, Jithitikulchai T, Hohmann J, Flessa S. Revisiting a decade of inequality in healthcare financial burden in Cambodia, 2009-19: trends, determinants and decomposition. Int J Equity Health. 2024;23:196. doi: 10.1186/s12939-024-02257-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Proportion of under-five illness episodes treated with antibiotics in Cambodia, by DHS year (2010, 2014, 2021–2022) and residence (urban/rural) for fever, ARI, and diarrhea